In recent posts we’ve talked about the market narrative, understanding where we are in the business cycle and the relative importance of certain asset classes over others as the economy ebbs and flows. Today, we’ll talk more about the actual process of using a market narrative to help make investment decisions. More importantly, we’ll show how we can use this process to potentially derive long term value in an investment portfolio.
Investing with the business cycle
One market narrative that we have been writing about is using ebbs and flows in the economy as a guide to an investment decision making process. The theory goes that when the economy is doing well, households save less and spend more. Factories ramp up production in response to inventory restocking and transportation companies increase deliveries to meet product needs. All this activity enables firms to generate more revenue and earnings on higher levels of spending.
Such virtuous economic times are positive for financial markets. This is because investors are more willing to take on risk in their portfolios and bid up asset prices in anticipation of higher future corporate earnings when there are expectations that the economy will do well. Further, stocks may outperform bonds during the good economic times while cyclical sectors of the market like consumer discretionary stocks or high yield bonds tend to outpace their defensive sector peers.
When the economic environment ahead looks challenged on the other hand, investors tend to seek out perceived safe-haven assets such as government bonds or defensive sectors like utilities or communications. Taken together, our market narrative suggests that an investor should therefore select a set of assets that align with upswings in the business cycle and another set of assets during a downswing.
Assumptions underpinnings our market narrative
Even so, how can we know with a degree of certainty that some assets tend to outperform others under different economic conditions? Further, how do we know that using the business cycle as a guide to investing can add value to a portfolio over the long term? Simple: test the assumptions.
Here are some assumptions flowing into our business cycle market narrative:
- Segmenting the business cycle into phases – the business cycle can be quantitatively segment into four phases (recovery, expansion, slowdown, recession).
- Phase-relevant asset performance – we can then use our quantitative phase identification work to select those assets that tend to perform better in the different phases of the economic cycle.
- Portfolio value add – finally, an investor can build a portfolio that outperforms a buy-and-hold strategy by rotating into and out of our selected asset groupings.
Segmenting the business cycle
We begin kicking the tires on our market narrative by taking a closer look at our first assumption: segmenting the business cycle. We do this to later objectively test the last two out of our three assumptions. It should be noted that the National Bureau of Economic Research (NBER) has done extensive work in the business cycle dating realm. Yet, their work mostly tells us about one phase of the business cycle (recessions) when we’re concerned with all four phases of the cycle.
What’s important in our work is being able to delineate between the different phase: when the end of a recession gives way to a recovery, an expansion, then to a slowdown just prior to the start of another recession. Currently there are few resources to assist with this phase identification other than the NBER’s recession identification work. Nevertheless, we have derived a method to quantitatively measure transitions between these important phases using our own business cycle indicator.
A visual representation of the cycle phases
The chart below, for example, represents our definition of the business cycle as defined by variations in the output gap. As a quick refresher, the output gap simply measures actual GDP growth versus how the economy should perform if it were clocking at full potential. In the chart, a value above the dotted line suggests the economy is producing above full potential which coincides with an expansion and later a slowdown, and a value below the line suggests that there is some slack in the economy following a recession and during a recovery.
We more precisely identify these four phases by looking at two key inflection points in the data. First, our cycle indicator historically tends to hit a low just after a recession prior to moving higher. We use an algorithm that applies a procedure that helps us confirm data is trending higher during an upswing and toward a recovery phase.
According to our measure, a recovery persists until the data passes above the dotted line which marks the end of a recovery and the start of an expansion. The second inflection point in the data comes at the end of an expansion when the data peaks and rolls over, marking the beginning of the slowdown phase. Again, we apply an algorithm to confirm this turning point. A recession thereafter is called when the indicator falls below zero (the dotted line), closing out the business cycle.
The results of our quantitative work applied to our preferred business cycle indicator are largely consistent with the historical business cycle dates used by the NBER to mark recessions. This suggests that we can indeed segment the business cycle in a consistent, repeatable manner. A fact that we’ll lean on in our next two tests.
Phase relevant asset performance
Now that we have a quantitative method to identify cycle phases, our next task is to validate our assumption that some assets perform better at different phases across the business cycle. For our test we selected a panel of 27 assets spanning stocks, bonds and sectors across various domestic and international markets.
We then calculated the historical quarterly price returns going as far back as 25 years for these assets and measured their performance during each of the four phases of the business cycle we identified in our prior step. Then we ranked the individual performance of the 27 assets within each the four business cycle phases.
What was the result of the test? Our work confirmed the general expectations investors have about asset performance and the economy. A top performing asset in one phase could be a bottom performing asset in another depending on where we are in the business cycle. For instance, U.S. government bonds outperformed during recessions in our work and U.S. large cap stocks during expansion phases which lines up with our professional experience about how the market tends to perform over time.
As a final point of validation, we computed the dispersion in returns of cyclical versus defensive assets to quantify the value add of changing rankings of asset classes during different phases of the cycle. To do this, we calculated the difference between top-ranked and bottom-ranked asset classes in each phase test.
What we find is a notable difference in the performance of the top and bottom 5-ranked asset classes which averages 7% per quarter over 25 years’ worth of data. This notable difference in best top versus bottom ranked asset classes further confirms our assumption that certain assets not only perform better (or worse) when the economy ebbs and flows, they outperform their peers by quite a margin.
Business cycle phases, asset selection and portfolio performance
We can now apply our new found knowledge about asset performance during certain phases of the business cycle within the context of an investment portfolio. The final assumption that we’re trying to address is whether one can add value to an investment portfolio by shifting into and out of assets that correspond with different phases of the business cycle.
Knowing whether such outperformance is possible requires us to create a portfolio whose asset weights adjust over time (tactical asset allocation) in response to the changing business cycle phases. We call this our cyclical portfolio. Then we perform a backtest using historical performance and economic data to determine whether our tactical overlay increases (or reduces) portfolio performance over time.
Backtesting our assumptions
To begin, we establish a baseline portfolio against which we can measure the performance of our tactical overlay. This portfolio uses a common 50/50 strategic asset allocation framework that is evenly divided between stocks and bonds and includes both allocations to foreign and domestic investments – a total of 11 assets.
Next, we define rules for our cyclical portfolio that enables us to lean into those assets we know perform well during a given phase of the business cycle. We set rules that dynamically adjust the weights of the 11 assets to account for business cycle phases, giving more weight to bonds during a slowdowns and recession and more weight to stocks during recoveries and expansions. From here we utilize roll-down rules that enable us to rotate into and out of the remaining 19 (out of 27) assets based on a given cycle phase in our backtest period.
Does the added effort of identifying the business cycle and selecting phase-relevant assets add value to a portfolio over time? Our findings suggest that the extra effort can add value. Using 15 years of historical return data the backtest results show that the cyclical portfolio outperforms a buy-and-hold portfolio by nearly a third over the 15 year backtest period.
Put differently, $100 invested 15 years ago would return $199 for our cyclical portfolio compared to $152 for the benchmark portfolio.
The value of a market narrative
Our work suggests that an investor can add value to their portfolio by identifying and selecting assets that perform well during certain phases of the business cycle.
Looking forward, a key challenge remains for investors seeking to apply such an investment strategy to their portfolio. That is, an investor not only needs to know where they are in the business cycle but know where it is headed in order to select assets that tend to perform well in a given cycle phase. This is no simple task and involves making a host of assumptions about the future.
To be sure, we talked about this important idea in one of our prior posts. The point is that a fundamental part of the market narrative process is understanding where the economy is headed and why. Having this understanding combined with a knowledge of which assets perform well under certain economic conditions we believe can potentially enable an investor to add some additional value to their investment portfolio.